Friday, November 28, 2008

Breaking Down Swing State Exit Polls: Campaign Contact

In this post, we are going to look through swing state exit poll data relating to whether voters were contacted by the respective campaigns.

National Exits

Contacted by any campaign? (2008 only)
Only Obama (13%) Obama 80, McCain 19
Only McCain (6%) McCain 82, Obama 17
Both (13%) McCain 51, Obama 47
Neither (66%) Obama 50, McCain 48

Contacted by the Obama/Kerry campaign?
Yes
(2004, 26%) Kerry 66, Bush 33
(2008, 26%) Obama 64, McCain 34

No
(2004, 74%) Bush 57, Kerry 42
(2008, 72%) McCain 50, Obama 48

Contacted by the McCain/Bush campaign?
Yes
(2004, 24%) Bush 62, Kerry 38
(2008, 18%) McCain 60, Obama 38

No
(2004, 76%) Kerry 52, Bush 47
(2008, 79%) Obama 55, McCain 43

Analysis. There is a wealth of information in this data which speaks very clearly to the effectiveness of the respective campaigns. Before looking over the information, we need to conceptualize how a good voter outreach operation works. In other words, who were the campaigns reaching out to, and why, and how does this data reveal whether the campaigns were effective in their outreach? In a campaign, the key group you are reaching out to is of course your base, the voters you want energized and out voting either on election day or earlier (depending on the state). After that, a good campaign is interested in reaching out to persuadables -- targeted voters who could swing to either side, and are thus worth lavishing attention on. Consider those voters only hit up by one campaign exclusively as base voters -- ones that one party wants to get out, and the other knows are unwinnable. The center is that group contacted by both parties, and is made up of many voters both campaigns see as winnable. Examining this latter group and the exits should give us a good idea which campaign was more persuasive with swingables.

Anyway, to my mind, one of the things that stands out most from this data is the wider net the Obama campaign was able to cast compared to that of Team McCain. As noted at the top, both campaigns had basically equal effectiveness with those voters who were only contacted by the individual campaigns, but the difference lies in the fact that there were twice as many such voters in Obama's pool than McCain's. Additionally, we can see that whereas Obama contacted the same percentage of voters as Kerry four years ago, McCain contacted 6% fewer voters than Team Bush in 2004. So, while Obama's level of success with contacted voters was slightly less than Kerry's, this was alright since the McCain campaign did such a mediocre job of voter contact.

Colorado

Contacted by any campaign? (2008 only)
Only Obama (22%) Obama 81, McCain 17
Only McCain (5%) ?
Both (29%) Obama 52, McCain 47
Neither (42%) McCain 53, Obama 45

Contacted by the Obama campaign?
Yes (51%) Obama 64, McCain 34
No (49%) McCain 56, Obama 42

Contacted by the McCain campaign?
Yes (34%) McCain 51, Obama 48
No (64%) Obama 57, McCain 41

Analysis. Colorado was one of the very most important swing states in this past election. He was highly coveted by both sides, and expected to be very close. In the end, Obama won an easy victory here in part because of his tremendous campaign in the state. Note that whereas Obama's campaign touched just over half of the state's voters, Team McCain only hit one-third of voters, an enormous difference, which, as you can see, translated to a wide disparity among ultimate vote totals. It is a sad commentary that the sample of size of voters contacted by only the McCain campaign was so small that we cannot discern how effective the campaign's core outreach was.

Florida

Contacted by any campaign? (2008 only)
Only Obama (15%) Obama 75, McCain 24
Only McCain (7%) McCain 71, Obama 27
Both (14%) McCain 52, Obama 45
Neither (63%) Obama 50, McCain 49

Contacted by the Obama/Kerry campaign?
Yes
(2004, 35%) Kerry 65, Bush 34
(2008, 29%) Obama 60, McCain 37

No
(2004, 65%) Bush 60, Kerry 39
(2008, 69%) McCain 51, Obama 48

Contacted by the McCain/Bush campaign?
Yes
(2004, 34%) Bush 63, Kerry 36
(2008, 20%) McCain 58, Obama 39

No
(2004, 66%) Kerry 54, Bush 44
(2008, 77%) Obama 55, McCain 44

Analysis. Again, we see that in terms of exclusive contacts, Obama's Florida operation hit twice as many voters as did McCain's, and the former's campaign was more effective to boot. Interestingly, far less voters were contacted in Florida than in 2004, by both sides. However, Team McCain only touched about one-fifth of voters in the state, a very lower number, and 14% below Bush/Cheney in 2004.

Indiana

Contacted by any campaign? (2008 only)
Only Obama (23%) Obama 73, McCain 26
Only McCain (8%) McCain 80, Obama 19
Both (14%) McCain 53, Obama 43
Neither (54%) McCain 53, Obama 46

Contacted by the Obama campaign?
Yes (37%) Obama 62, McCain 36
No (62%) McCain 57, Obama 43

Contacted by the McCain campaign?
Yes (22%) McCain 63, Obama 34
No (77%) Obama 54, McCain 45

Analysis. As you can see, Indiana was a state where voter outreach made a huge difference. Over the summer, we postulated that McCain would have to leave states like Indiana untouched in hopes that they could hold on without large infusions of money. The rationale behind this view was simply that Team McCain would not have the funds to fight in traditional red states. In the end, McCain's (necessary) gambit failed, as Obama poured a lot of resources into Indiana with great impact. Obama hit 15% more voters than McCain. Obama lost every cross-section by a health margin except those nearly one-quarter of voters who were contacted. Obama won this group by just enough to win statewide. Voter outreach here made a big difference. Indeed, McCain badly lost voters he never contacted.

Nevada

Contacted by any campaign? (2008 only)
Only Obama (26%) Obama 80, McCain 18
Only McCain (5%) McCain 88, Obama 10
Both (24%) Obama 50, McCain 49
Neither (43%) Obama 50, McCain 47

Contacted by the Obama campaign?
Yes (50%) Obama 66, McCain 33
No (48%) McCain 51, Obama 46

Contacted by the McCain campaign?
Yes (29%) McCain 55, Obama 43
No (69%) Obama 62, McCain 36

Analysis. I've said it before, and in light of this data I will say it again: the individuals who ran Obama's Nevada operation should be taken care of in the administration. Just look at one number: 50, as in the percentage of voters who were contacted by the Obama campaign. Half of the state's voters received a contact, and of that bloc, Obama won two out of every three votes. That's your ballgame right there. Now, Nevada was seemingly predisposed to support Obama this year, as indicated by McCain's slim lead among voters Obama did not contact (as opposed to a wide lead, which would have been expected). Still, Obama's incredible state operation played a big role in shaping that predisposition. How the heck did Obama exclusively contact one-quarter of voters, while McCain similarly hit only 5% in a state the GOP had only lost twice since 1948? I guess it was true what the NY Times reported when an unnamed operative said that McCain had no presence here.

North Carolina

Contacted by any campaign? (2008 only)
Only Obama (14%) Obama 89, McCain 10
Only McCain (6%) McCain 85, Obama 15
Both (20%) Obama 50, McCain 49
Neither (58%) McCain 55, Obama 44

Contacted by the Obama campaign?
Yes (34%) Obama 66, McCain 33
No (64%) McCain 58, Obama 41

Contacted by the McCain campaign?
Yes (26%) McCain 58, Obama 41
No (72%) Obama 53, McCain 46

Analysis. Because the two campaigns split down the middle the pool of switchables contacted by both camps, this race came down to which side was able to bring out its base better, and on that score, Obama won easily. He again doubled up on base contacts here, and brought them to the polls at a better clip.

Ohio

Contacted by any campaign? (2008 only)
Only Obama (15%) Obama 81, McCain 19
Only McCain (8%) McCain 82, Obama 18
Both (28%) McCain 51, Obama 48
Neither (46%) Obama 51, McCain 48

Contacted by the Obama campaign?
Yes (43%) Obama 59, McCain 40
No (54%) McCain 53, Obama 46

Contacted by the McCain campaign?
Yes (36%) McCain 58, Obama 41
No (61%) Obama 58, McCain 41

Analysis. It seems like both sides were effective in turning people they contacted, but Obama contacted more people.

Virginia

Contacted by any campaign? (2008 only)
Only Obama (22%) Obama 81, McCain 19
Only McCain (10%) McCain 84, Obama 15
Both (28%) McCain 57, Obama 42
Neither (38%) Obama 53, McCain 46

Contacted by the Obama campaign?
Yes (50%) Obama 59, McCain 40
No (48%) McCain 54, Obama 45

Contacted by the McCain campaign?
Yes (38%) McCain 64, Obama 35
No (59%) Obama 63, McCain 36

Analysis. Like in Nevada, the Obama campaign was able to reach out to half of the people who ultimately voted. This is a hard thing to beat. Interestingly, Obama badly lost voters targeted by both sides, but he got his winning margin by comfortably winning uncontacted voters.

No comments: